Review Article

J. Fluids Eng. 2017;139(3):030801-030801-21. doi:10.1115/1.4035116.

This paper is a historical review of the science, both experimental and theoretical, behind the iconic Moody diagram used to avoid tedious iterations choosing pumps and pipes. The large body of historical pipe flow measurements and the choice of dimensionless groups and the Buckingham-Π theorem are also discussed. The traditional use of the Moody diagram to solve common pipe flow problem is discussed. Alternatives to the Moody diagram from the literature and novel ones presented here are shown to produce a solution without iteration for any type of pipe loss problem.

Commentary by Dr. Valentin Fuster

Research Papers: Flows in Complex Systems

J. Fluids Eng. 2016;139(3):031101-031101-14. doi:10.1115/1.4035026.

Understanding the formation mechanism of the S-shaped characteristics (SSCs) and the relationship between flow structures and the runaway instability (RI) is the prerequisite for optimizing runner design to promote operational reliability and flexibility. In this study, a new turbine equation is derived to reveal the prime cause of the SSCs, and the influence of geometric parameters on the SSCs is analyzed. Moreover, the flow patterns in three model turbines of different specific-speeds are simulated by unsteady Computational fluid dynamics (CFD), and the correlation between inverse flow vortex structures (IFVSs) and the RI in the SSCs region is identified. Theoretical analysis shows that the turbine equation can theoretically predict the change trend of the first quadrant SSCs curves of the pump-turbines; the flow losses caused by small blade inlet angle, instead of the diameter ratio, are the primary cause of the SSCs. The numerical simulation results reveal that the IFVSs at the hub side of the runner inlet are the origin of the RI; when operating points are far away from the best efficiency point (BEP), the IFVS locations change regularly. For large guide vane openings (GVOs), the IFVSs first incept at the shroud side, and then translate to the hub side, and further back to the midspan, when the discharge decreases. The inception points (IPs) of the SSCs correspond to the onset of the IFVSs at the hub side, which are in advance of the zero-torque operating points (ZTOPs); therefore, the ZTOPs are located in the positive slope region, leading to RI. For small GVOs, however, the IFVSs only locate at the midspan; the IPs of the SSCs, having no definite correlation with the IFVSs, are coincided with or are below the ZTOPs, because the ZTOPs are in the negative slope region and RI disappears. It is also found that the IPs of SSCs are the turning points of the predominant states between the turbine effect and pump effect. These results are valuable for design and optimization of pump-turbine runners.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2016;139(3):031102-031102-12. doi:10.1115/1.4034864.

An experimental and numerical investigation was carried out to explore the effects of four vortex generators (VG) on the onset of flow instabilities, the paths and characteristics of the induced coherent counter-rotating vortices at a Reynolds number Re ≈ 2000. The flow field around the VG was characterized using a smoke visualization technique and simulated numerically using Reynolds-averaged Navier-Stokes (RANS). The taper angle of the VG was varied based on the used tab geometries, including triangular, trapezoidal, and rectangular tabs, which shared the same height, inclination angle, and base width. The results reveal that each VG was able to generate a counter-rotating vortex pair (CVP), and that the taper angle has direct effects on the path of the CVP, the onset location of Kelvin–Helmholtz (K-H) instabilities, and the circulation strength of the vortex structures. Furthermore, a linear relation between VG taper angle and the onset of instability was observed experimentally. Before the onset of K–H instability, the path of the CVP in the wake of a VG can be predicted using a pseudo-viscous model, which was validated numerically and experimentally.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2017;139(3):031103-031103-11. doi:10.1115/1.4034951.

Influences of proper fence installation around solar farms for decreasing aerodynamic factors due to wind force on parabolic trough collectors are comprehensively studied using two-dimensional computational fluid dynamics (CFD) model. Fences are treated as porous media to be investigated from the viewpoint of their influences on wind flow. The aerodynamic factors are calculated for the collectors in case of different fence types. Comprehensive discussions about the effects of types of employed fences and their distance from the first row collectors as well as collectors' slope angle on aerodynamic forces are also presented. Sheltering and fence effects are considered by an innovative modeling approach that is proposed in this study. It is shown that fence installation can considerably decrease aerodynamic factors. Effects of formed vortices behind collectors were significant and should be taken into consideration during the design. Brick-type fences are shown to behave poor while lace-type fences are advised.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2017;139(3):031104-031104-11. doi:10.1115/1.4034876.

A general hydraulic loss coefficient correlation for perpendicular, cylindrical, finite length dividing pipe junctions is developed and implemented in a discrete dividing-flow manifold model. Dividing-flow manifolds are used in several technical appliances, e.g., in water and wastewater treatment, swimming pool technology, air engineering, and polymer processing. Ensuring uniform flow distribution is a major goal of a flow manifold system design, whose accuracy is usually determined by the accuracies of applied flow coefficients. Coefficient of turning losses is calculated by a computational fluid dynamics (CFD)-based approach applying a nonlinear fit. In the case of a single-phase flow, the loss coefficient depends on four dimensionless parameters: the Reynolds number, the ratio of port and header flow velocities, the diameter ratio, and the ratio of the port length and the diameter of the pipe. Instead of experimentally covering this four-dimensional parameter space, more than 1000 judiciously chosen three-dimensional simulations were run to determine the loss coefficient for the parameter range most used in engineering practice. Validated results of our novel resistance formula show that the velocity and port length to header diameter ratios have a significant effect on the turning loss coefficient, while the diameter ratio and Reynolds number dependency are weaker in the investigated parameter ranges.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2017;139(3):031105-031105-15. doi:10.1115/1.4035137.

A very detailed experimental case of a reversed profile in ground effect has been selected in the open literature where available experimental data have been used as reference data for the computational fluid dynamics (CFD) analysis. The CFD approach has been used to predict aerodynamic performance of the profile at different distances with respect to the ground: in the freestream case, there is no ground effect whereas in the low height the profile operation is limited by the stall on the suction surface. Moreover, the effect of a Gurney flap addition on flow distribution and performance has been numerically investigated. The experimental data have been used to setup and test the capabilities of the computational approach. With the addition of a Gurney flap, a significant flow unsteadiness is introduced that needs to be considered in the numerical approach. In this case, the configurations investigated are used to highlight the capabilities of CFD using Reynolds-averaged Naiver–Stokes (RANS) approach for its effective application as a tool for the detailed design of aerodynamic components to generate downforce for race cars.

Commentary by Dr. Valentin Fuster

Research Papers: Fundamental Issues and Canonical Flows

J. Fluids Eng. 2017;139(3):031201-031201-8. doi:10.1115/1.4035012.

In the present study, numerical investigation is carried out for flow past a transversely oscillating circular cylinder near a wall. A second-order finite-volume method employing diffuse interface immersed boundary method is used to handle the nonconforming boundaries. The cylinder is allowed to oscillate on a fixed Eulerian mesh in order to handle the moving boundary. Simulations are carried out for a number of forcing frequencies at three different gap distances and two amplitudes of oscillation with Reynolds number fixed at 200. While a pair of vortices is found to be shed near the stationary shedding frequency at larger gap distance, multiple interconnected vortices are observed at larger forcing frequencies. Proximity of the wall promotes greater interaction of the wall layer with the near wake resulting in inhibited irregular shedding. Energy transfer changes its direction when the correlation between the cylinder motion and the lift force is strongest. Positive energy transfer attains a peak at the onset of the synchronization regime accompanied by a weaker correlation. Amplitude of oscillation of the cylinder evidently has systematic effect on the drag coefficient and wake fluctuations, though lift force remains grossly unaltered. Lower amplitude of motion favors induced motion as opposed to larger ones where greater negative energy transfer occurs.

Commentary by Dr. Valentin Fuster

Research Papers: Multiphase Flows

J. Fluids Eng. 2017;139(3):031301-031301-11. doi:10.1115/1.4034952.

Previous work has shown that the employment of a gap drainage impeller in a centrifugal pump can improve the pump's hydraulic performance and cavitation resistance. However, during experiments, an unconventional cavitation phenomenon has been observed in the form of a staggered pair of fixed impeller flow tunnels. For the purpose of understanding the factors involved with this unconventional phenomenon, the present study analyzes the cavitation formation and evolution processes using numerical and experimental methods. A scalable detached eddy simulation (SDES) method was employed to address unsteady turbulent flow. First, the method was validated by comparing the performance data and liquid water velocity distributions obtained by calculation and experiment in the absence of cavitation. Then, numerical simulations of the cavitation flow field were conducted under a flow discharge condition one-half that of the optimum value. Within a particular range of cavitation numbers, the calculated results are found to reproduce the unconventional cavitation phenomenon observed in the experiments. The formation mechanism involves a combination of many factors such as impeller geometry, inflow discharge condition, and cavitation number. As for a certain geometry, the formation and evolution processes can generally be analyzed and explained according to the influence of the attack angle, which is affected by variations in the allocated flow discharge and cavitation volume in each impeller tunnel. The jet flow through the gap between the main and vice blades also contributes to the formation of this phenomenon.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2017;139(3):031302-031302-12. doi:10.1115/1.4034758.

Pressure drop has been measured for upward single- and two-phase gas–liquid flow across an orifice in a vertical pipe. A conductance probe provided average void fraction upstream of the orifice. Six orifices with different apertures/thickness were mounted in turn in a 34 mm diameter transparent acrylic resin pipe. Gas and liquid superficial velocities of 0–4 m/s and 0.3–0.91 m/s, respectively, were studied. For single-phase flow, pressure drop, expressed as an Euler number, was seen to be independent of Reynolds number in turbulent region. The Euler number increased with decreasing the open area ratio/orifice thickness and increasing velocity. The pressure drop was well predicted by the correlation of Idel'chik et al. (1994, Handbook of Hydraulic Resistances, 3rd ed., CRC Press, Boca, Raton, FL.), which uses a form of Euler number. The corresponding two-phase flow pressure drop depends on the flow pattern. Decreasing open area ratio/orifice thickness increased the pressure drop. For a given liquid superficial velocity, the pressure drop increases with gas superficial velocity except for low open area ratio where this increase is followed by a decrease beyond a critical superficial gas velocity for the high liquid superficial velocities. Relevant correlations were assessed using the present data via a systematic statistical approach. The two-phase multiplier equations of Morris (1985, “Two-Phase Pressure Drop Across Valves and Orifice Plates,” European Two Phase Flow Group Meeting, Marchwood Engineering Laboratories, Southampton, UK.) and Simpson et al. (1983, “Two-Phase Flow Through Gate Valves and Orifice Plates,” International Conference on Physical Modelling of Multiphase Flow, Coventry, UK.) are the most reliable ones.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2017;139(3):031303-031303-10. doi:10.1115/1.4035013.

Experimental results of the void fraction, statistical chord length distribution (CLD), and bubble size distribution (BSD) inside and downstream of hydrodynamic cavities are presented at the laboratory scale. Various cavitating flows have been intensively studied in water tunnels for several decades, but no corresponding quantitative CLD and BSD data were reported. This experimental study is aimed at elaboration of a general approach to measure CLD in typical cavitating flows. Dual-tip electrical impedance probe (dtEIP) is used to measure the void fraction and CLD in different cavitation flows over a flat plate, including both supercavitation and sheet/cloud cavitation. For supercavitating flows, the void fraction of vapor is unity in the major cavity region. In contrast, the maximum void fraction inside the sheet/cloud cavitation region is less than unity in the present studies. The high vapor concentration region is located in the center of the cavity region. Based on the experimental data of CLD, it is found that the mean chord lengths are around 2.9–4.8 mm and 1.9–4.4 mm in the center region and closure region, respectively. The backward converting bubble diameters at the peak of BSD have similar magnitude, with probability density values exceeding 0.2. Empirical parameters of CLD and BSD are obtained for different cavity regions.

Commentary by Dr. Valentin Fuster

Research Papers: Techniques and Procedures

J. Fluids Eng. 2016;139(3):031401-031401-3. doi:10.1115/1.4034950.

The standard expression for pipe friction calculations, the Colebrook equation, is in an implicit form. Here, we present two accurate approximate solutions, given by replacing the numerically unstable term in Keady's exact Lambert function solution with a truncated series expansion. The resulting expressions have a higher accuracy than most advanced approximations and a lower computational cost than basic engineering formulas. The simplest expression, given by retaining only three terms in the series expansion, has a maximum error of less than 0.153% for Re ≥ 4000. The slightly more involved expression, based on five terms, has a maximum error of 0.0061%.

Topics: Approximation , Errors
Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2017;139(3):031402-031402-11. doi:10.1115/1.4034953.

Sculpting inertial fluid flow using sequences of pillars is a powerful method for flow control in microfluidic devices. Since its recent debut, flow sculpting has been used in novel manufacturing approaches such as microfiber and microparticle design, flow cytometry, and biomedical applications. Most flow sculpting applications can be formulated as an inverse problem of finding a pillar sequence that results in a desired fluid transformation. Manual exploration and design of pillar sequences, while useful, have proven infeasible for finding complex flow transformations. In this work, we extend our automated optimization framework based on genetic algorithms (GAs) to rapidly design micropillar sequences that can generate arbitrary user-defined fluid flow transformations. We design the framework with the following properties: (a) a parameter encoding that respects locality to ensure fast convergence and (b) a multiresolution approach that accelerates convergence while maintaining accuracy. The framework also utilizes graphics processing unit (GPU) architecture via NVIDIA's CUDA for function evaluations. We package this framework in a user-friendly and freely available software suite that enables the larger microfluidics community to utilize these developments. We also demonstrate the framework's capability to rapidly design arbitrary fluid flow shapes across multiple microchannel aspect ratios.

Commentary by Dr. Valentin Fuster

Technical Brief

J. Fluids Eng. 2017;139(3):034501-034501-6. doi:10.1115/1.4035138.

This paper presents new models of entropy production for incompressible turbulent channel flows. A turbulence model is formulated and analyzed with direct numerical simulation (DNS) data. A Reynolds-averaged Navier–Stokes (RANS) approach is used and applied to the turbulence closure of mean and fluctuating variables and entropy production. The expression of the mean entropy production in terms of other mean flow quantities is developed. This paper presents new models of entropy production by incorporating the eddy viscosity into the total shear stress. Also, the Reynolds shear stress is used as an alternative formulation. Solutions of the entropy transport equations are presented and discussed for both laminar and turbulent channel flows.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In